基于SWAT模型的流域侵蚀产沙影响因子研究

发布时间：2018-03-16 23:03

本文选题：SWAT模型　切入点：侵蚀产沙　出处：《华北电力大学(北京)》2016年硕士论文　论文类型：学位论文

【摘要】：土壤侵蚀是引起土地退化和河流泥沙淤积的根源。系统研究流域侵蚀产沙变化与降雨、土壤、土地利用、地貌条件等影响因子之间的关系,获得不同类型区域在不同影响因子下的侵蚀产沙规律,不仅可以加强区域水土保持措施的针对性,而且可以加深对地球物理过程的认识,提高土地变化科学研究的理论水平。流沙河流域作为川西南典型的生态环境脆弱区之一,水土流失严重、生态环境破碎化,干旱、洪涝、滑坡、泥石流等山地灾害频繁发生,成为区域经济社会发展的主要限制性因素。本文以流沙河流域为研究对象,选用基于ArcGIS的SWAT分布式水沙模型,针对不同土地利用情景下不同降雨强度、不同土壤类型的流域侵蚀量进行模拟,结合流域景观格局指标变化量计算,研究了流域侵蚀产沙与各影响因子之间的关系,主要研究内容与结论如下：(1)建立了流沙河流域分布式水沙模型,采用1980-2006年实测径流、泥沙资料对模型进行了率定验证。率定验证结果(月径流量Ens0.71, R20.95,月泥沙量Ens0.75, R20.87)表明模型具有较好的精度,可以用于探究多因素影响下的水沙产输特性。采用退耕还林和退草还耕两种土地利用变化情景作为模型的输入,进行土地利用变化在不同降雨强度与不同土壤下垫面下的产沙量情景模拟。结果表明：退草还耕和退耕还林两种土地利用变化下流域产沙对降雨强度的敏感性退草还耕要远大于退耕还林；不同士壤类型下,无论是退耕还林还是退草还耕,下垫面为黄土时土地利用变化造成的侵蚀量相对变化最大,以下依次是黑土、红壤和紫色土；而不同土壤类型在两种土地利用变化下的产沙量对降雨强度的敏感性则相反,对降雨强度最敏感的是紫色土,其次是红壤土、黑土和黄土。(2)以子流域为样本,利用Fragstats软件分别计算1986年、1995年、2000年三期土地利用下各个子流域的景观指标,建立了各个子流域景观格局指标变化量与子流域侵蚀产沙模数变化量的对应关系,采用基于最小二乘法的多元线性回归方法进行统计分析,筛选出影响流域侵蚀量变化的4个典型景观格局指标,并得到了流沙河流域土壤侵蚀模数变化量与典型景观格局指标变化量的多元回归方程。根据回归方程中各变量的实际物理意义及相应景观指标含义,分析了流域侵蚀产沙与部分景观指标的相关性,为流域水土保持、土地利用景观优化管理等工作提供理论参考。
[Abstract]:Soil erosion is the root cause of land degradation and river siltation. The regularity of erosion and sediment yield in different types of regions under different influence factors can not only strengthen the pertinence of regional soil and water conservation measures, but also deepen the understanding of geophysical processes. Improving the theoretical level of scientific research on land change. As one of the typical fragile areas of ecological environment in southwest Sichuan, the Liu Shahe basin is characterized by serious soil and water loss, fragmentation of ecological environment, drought, flood, landslide, debris flow, and other mountain disasters. This paper takes Liu Shahe watershed as the research object, chooses the SWAT distributed water and sediment model based on ArcGIS, aiming at different rainfall intensity under different land use scenarios. The erosion amount of different soil types was simulated, and the relationship between erosion and sediment yield and the influencing factors was studied based on the calculation of watershed landscape pattern index. The main research contents and conclusions are as follows: (1) A model of distributed water and sediment in Liu Shahe watershed is established, and the runoff measured from 1980 to 2006 is used. The results of rate determination (monthly runoff Ens0.71, R20.95, monthly sediment quantity Ens0.75, R20.87) show that the model has good accuracy. It can be used to study the characteristics of water and sediment yield and transportation under the influence of many factors. Two land use change scenarios, namely returning farmland to forest and returning grass to tillage, can be used as the input of the model. A scenario simulation of land use change under different rainfall intensity and under different soil underlying surface was carried out. The results showed that the sensitivity of sediment yield to rainfall intensity was reduced under the two land use change conditions, I. e., returning grass to tillage and returning cropland to forest. The return of grass to farming is much greater than returning cropland to forest; Under different soil types, the erosion amount caused by the change of land use was the biggest when the underlying surface was loess, followed by black soil, red soil and purple soil. The sensitivity of sediment yield of different soil types to rainfall intensity was opposite, purple soil was the most sensitive to rainfall intensity, followed by red loam soil, black soil and loess. The landscape indexes of each sub-watershed under land use in 1986, 1995 and 2000 were calculated by Fragstats software, and the corresponding relationship between the change of landscape pattern index and the modulus of erosion and sediment yield in each sub-watershed was established. The multivariate linear regression method based on least square method was used for statistical analysis, and four typical landscape pattern indexes affecting the change of watershed erosion were selected. The multivariate regression equations of soil erosion modulus variation and typical landscape pattern index change in Liu Shahe watershed are obtained. According to the actual physical meaning of each variable and the corresponding landscape index meaning in the regression equation, the soil erosion modulus variation and the typical landscape pattern index change are obtained. The correlation between erosion and sediment yield and some landscape indexes is analyzed, which provides a theoretical reference for watershed soil and water conservation, land use landscape optimization management and so on.
【学位授予单位】：华北电力大学(北京)
【学位级别】：硕士
【学位授予年份】：2016
【分类号】：S157

【相似文献】